European Heart Journal (2002) 23, 13291344

doi:10.1053/euhj.2002.3274, available online at http://www.idealibrary.com on

Task Force Report

Guidelines for the interpretation of the neonatal
electrocardiogram
A Task Force of the European Society of Cardiology
P. J. Schwartz1 (Chair), A. Garson, Jr2, T. Paul3, M. Stramba-Badiale4,
V. L. Vetter5, E. Villain6 and C. Wren7
1

Department of Cardiology, University of Pavia and IRCCS Policlinico S. Matteo, Pavia, Italy; 2University of
Virginia, Charlottesville, VA, U.S.A.; 3The Childrens Heart Program of South Carolina, Medical University of
South Carolina, Charleston, SC, U.S.A.; 4Pediatric Arrhythmias Center, IRCCS Istituto Auxologico Italiano,
Milan, Italy; 5Division of Pediatric Cardiology, Department of Pediatrics, Childrens Hospital of Philadelphia,
University of Pennsylvania School of Medicine, Philadelphia, PA, U.S.A.; 6Division of Pediatric Cardiology,
Department of Pediatrics, Hopital Necker Enfants Malades, Paris, France; 7Department of Paediatric Cardiology,
Freeman Hospital, Newcastle upon Tyne, U.K.

Introduction.............................................................1329
Normal electrocardiogram in the newborn .............1330
Normal values......................................................1330
Technology ..........................................................1330
Artefacts...............................................................1332
Electrocardiographic measurements ....................1332
Heart rate.............................................................1332
P wave..................................................................1332
QRS complex.......................................................1332
QT interval ..........................................................1333
ST segment and T wave ......................................1333
Abnormal electrocardiogram in the newborn .........1333
Heart rate.............................................................1333
Sinus arrhythmia..............................................1333
Sinus tachycardia .............................................1333
Sinus bradycardia.............................................1335
Other bradycardias...........................................1335
P wave..................................................................1335
Atrioventricular conduction.................................1335
Complete (3rd) atrioventricular block .............1335
1st and 2nd atrioventricular block...................1336
Intraventricular conduction .................................1336
Bundle branch block ........................................1336
Non-specific intraventricular conduction
abnormalities....................................................1336
WolParkinsonWhite syndrome ..................1336
QRS axis and amplitude......................................1338
Right ventricular hypertrophy .........................1338

Correspondence: Peter J. Schwartz, MD, FESC, FACC,, FAHA,

Professor & Chairman, Department of Cardiology, Policlinico S.
Matteo IRCCS, Viale Golgi, 19-27100 Pavia, Italy.
0195-668X/02/$35.00

Left ventricular hypertrophy............................1338

Low QRS voltage.............................................1338
Ventricular repolarization....................................1338
QT prolongation: dierential diagnosis ...........1339
Long QT syndrome..........................................1339
ST segment elevation .......................................1341
Atrial and ventricular arrhythmias......................1341
Atrial/junctional ...............................................1341
Premature atrial beats ..................................1341
Supraventricular tachycardia........................1342
Atrial flutter .................................................1342
Ventricular arrhythmias ...................................1342
Premature ventricular beats..........................1342
Ventricular tachycardia ................................1343
Accelerated ventricular rhythm ....................1343
Conclusion...............................................................1343
Acknowledgements ..................................................1343
References................................................................1343

Introduction
Most cardiologists who care for adults have no
or minimal experience with electrocardiograms
(ECGs) recorded in infants. So far, this has had no
practical implications because only seldom are they
requested to examine a neonatal ECG. This situation,
however, may change as some European countries
have begun to consider the possibility of introducing in their National Health Services the performance
of an ECG during the first month of life in all
newborns, as part of a cardiovascular screening
programme.

 2002 Published by Elsevier Science Ltd on behalf of The European Society of Cardiology

1330

Task Force Report

The background of this evolution is multiple, but it

lies largely in the realization that early identification
of life-threatening arrhythmogenic disorders, which
often manifest in infancy, childhood or even later,
may allow initiation of eective preventive therapy. A
large prospective study has indicated that some infants with prolonged QT interval in the first week of
life had sudden death, and would have previously
been labelled as victims of the Sudden Infant Death
Syndrome[1]. Furthermore, in infants with this diagnosis, post-mortem molecular screening may reveal
the presence of the long QT syndrome (LQTS)[2]. As
with most screening tests, a single ECG must be put
into context (e.g. family history, etc.). Additionally, it
is traditional to examine neonatal ECGs looking for
those with parameters below the 2nd or exceeding
the 98th percentile. While it is true that these
values are abnormal in a strict statistical sense, very
often abnormality does not imply the presence of a
disease, or of a risk for clinically relevant events. This
depends largely on the parameter under examination.
However, also the reverse may be true and, in the
neonate, a completely normal ECG may be seen with
multiple types of congenital heart defects and with the
entire spectrum of arrhythmias. This call for caution
does not detract from the valid concept that the
identification of ECG abnormalities in the newborn
can be the first step toward a meaningful act of
preventive medicine.
Should this neonatal screening indeed be introduced as part of National Health Services, then hospital cardiologists most of whom are unfamiliar
with neonatal ECGs would be asked to read these
tracings. The European Society of Cardiology (ESC)
has realized the potential implications for European
cardiologists and for health care, and has acted accordingly. Through the Committee for Practice
Guidelines and Policy Conferences, chaired by
Werner Klein, it has instituted this Task Force. The
experts were designated by the Guidelines Committee
and approved by the Board of the ESC. The panel
was composed of physicians and scientists involved in
clinical practice in University and non-University hospitals. Members were selected to represent experts of
dierent European countries; in addition, two nonEuropean members were included for their worldwide
recognized expertise in the field of pediatric electrocardiography.
The ESC considers medical education and the
improvement of clinical practice among its major
obligations. The main objective of the present report
is to present adult cardiologists with a consensus
document designed to provide guidelines for the
interpretation of the neonatal ECG, focusing on
the most clinically relevant abnormalities and on the
ensuing management and referral options. This
document aims also at providing paediatricians and
neonatologists with updated information of clinical
relevance that can be detected from a neonatal
ECG.
Eur Heart J, Vol. 23, issue 17, September 2002

The procedure used for developing and issuing these

guidelines was in accordance with the recently issued
Recommendations for Task Force creation and report
writing,
(http://www.escardio.org/scinfo/guidelines_
recommendations.htm) which is a position document of
the ESC Committee for Practice Guidelines and Policy
Conferences.
This document was reviewed and approved by the
Commitee for Practice Guidelines and Policy Conferences. It was endorsed by the Board of the ESC and
represents the ocial position of the ESC with regard to
this subject. These guidelines will be reviewed two years
after publication and considered as current unless the
Guidelines Committee revises or withdraws them from
circulation.
This Task Force was financed by the budget of the
Committee for Practice Guidelines and Policy Conferences of the ESC and was independent of any commercial, health or governmental authorities.

Normal electrocardiogram in the

newborn
Normal values
Changes occur in the normal ECG from birth to adult
life. They relate to developmental changes in physiology,
body size, the position and size of the heart relative to
the body, and variations in the size and position of the
cardiac chambers relative to each other. The major
changes in the paediatric ECG occur in the first year of
life with the majority of normal adult values being
abnormal in the newborn. Likewise, many normal newborn values and patterns would be abnormal in the
adult. Normal electrocardiographic values in the paediatric population traditionally derive from those published in 1979 by Davignon et al.[3]. From the ECGs of
1027 infants less than 1 year of age and among these, 668
in the first month of life, the percentile distribution of
electrocardiographic variables was calculated. It is important to refer to tables of normal values as shown in
Table 1. A recent large study by Rijnbeek et al.[4]
included an extremely low number of neonates (n=44)
but no one below 3 weeks of age. Thus, the percentile
tables published by Davignon are recommended for use
in clinical practice. Other references on the reading of
ECGs in neonates and children are available[5,6].

Technology
The normal newborn ECG should include 12 leads.
Other leads, V3R, V4R and V7, may provide additional
information to evaluate possible congenital heart
lesions.
The current use of computerized digital ECG systems
aects newborn ECGs to a greater extent than those of
older children or adults[7]. The newborn ECG may have

93154 (123)
91159 (123)
90166 (129)
107182 (149)
121179 (150)

01 days
13 days
37 days
730 days
13 months

+59
+64
+77
+65
+31

to
to
to
to
to

+192 (135)
+197 (134)
+187 (132)
+160 (110)
+114 (75)

Frontal plane
QRS axisa
(degrees)

From ref. [3].

a
2nd98th percentile (mean)
b
2nd98th percentile (1 mm=100
V)
c
98th percentile (1 mm=100
V)

Heart rate
(beats . min
1)

Normal neonatal ECG standards*

Age group

Table 1

28
28
29
30
26

P wave
amplitude
(mm)
008016 (011)
008014 (011)
008014 (010)
007014 (010)
007013 (010)

P-R
intervala
(s)
002008 (005)
002007 (005)
002007 (005)
002008 (005)
002008 (005)

QRS
durationa
V5
52
52
48
56
54

17
21
28
28
27

Q IIIc QV6c
(mm) (mm)

526
527
324
3215
3185

RV1b
(mm)

0225
021
017
011
0125

SV1b
(mm)

98
6
97
7
74

R/S V1c

011
012
0512
2516
521

RV6b
(mm)

098
095
098
098
072

10
11
10
12
12

28
29
25
22
29

52
52
48
47
53

SV6b
SV1 +RV6c R+SV4c
R/S V6c
(mm)
(mm)
(mm)

Task Force Report

1331

Eur Heart J, Vol. 23, issue 17, September 2002

1332

Task Force Report

a higher voltage and shorter duration QRS complexes

resulting in a higher percentage of high frequency components. The recommendations of a number of groups
vary as to the best bandwidth cutos and sampling
frequency to reduce error[8,9]. Higher bandwidth cutos
may alter amplitude of signals by as much as 46%[10].
This would make standards determined from analogue
signals or digitized signals at lower sampling rates and
lower frequency cutos dierent from those at higher
settings. The current American Heart Association recommendation for paediatric ECGs is 150 Hz as a minimum bandwidth cuto and 500 Hz as a minimum
sampling rate[11]. The Rijnbeek study reported normals
using a higher sampling rate of 1200 Hz. Compared to
Davignons study, which used a sampling rate of 333 Hz,
the newborn upper limits in Rijnbeeks study were
1225% higher than in Davignons[3,4].

Artefacts
Artefacts are common in newborn ECGs and include
limb lead reversal and incorrect chest lead positioning.
In addition, electrical interference, usually 60 cycles,
can occur in hospital settings from bedside monitors,
warmers or other equipment.
Other artefacts occur because of various types of
patient movement common in neonates. These artefacts
may be random as with hiccoughs or limb movement.
Normal complexes are seen along with the artefacts, and
the intrinsic rhythm of the patient is not aected. Other
common artefacts include a fine, often irregular undulation of the baseline from muscle tremors or jitteriness.
Again, the intrinsic rhythm is not aected. The size of
the QRS complex and the baseline may wander in a
cyclic fashion with respirations. It should be noted that
the neonate breathes from 3060 times per min.
The main clue in determining the presence of an
artefact is to evaluate whether it aects the intrinsic
rhythm and if it is timed such that it could be a true
depolarization. A signal within 80 ms from a true QRS
complex could not occur from an electrophysiologic
point of view.

Electrocardiographic measurements
Because of the current limitations of electronic measurements in newborn ECGs, intervals should be hand
measured as the computerized systems are often inaccurate in the newborn. Intervals in children increase with
increasing age, reaching most of the adult normal values
by 78 years of age.

Heart rate
Heart rate can be determined by a variety of methods. It
should be noted that normal neonates may have rates
Eur Heart J, Vol. 23, issue 17, September 2002

between 150230 beats . min
1, especially if they are

crying or agitated. Over 200 beats . min
1, one-half
small box can make an appreciable dierence in heart
rates.
Heart rates between the 2nd and 98th percentile in the
first year of life are shown in Table 1. The normal heart
rate increases from the first day of life, it reaches a peak
between the first and the second month and then declines returning to the values recorded at birth by the
sixth month. During the following 6 months, it remains
rather stable and then slowly declines after 1 year due to
maturation of vagal innervation of the sinus node[12].
Clinically significant gender dierences in heart rate are
not seen in the neonatal period.

P wave
The P wave axis is a vector indicating the direction of
activation, which is away from the site of origin. By
identifying the quadrant location of the P wave axis one
can determine the site of origin of the rhythm. For
example, sinus rhythm originates in the high right
atrium transcribing a P wave with an axis in the quadrant bordered by 0 and +90. Measurements are available for P wave amplitude (Table 1). The P wave is
generally pointed in lead II and aVF and more rounded
in other leads. Lead V1 may be diphasic.
The PR interval is measured from the onset of the P
wave to the Q or R wave if no Q wave is present. The PR
interval, measured in lead II, increases with age and
decreases with heart rate. The normal neonatal PR
interval ranges from a minimum of 70 ms to a maximum
of 140 ms, with a mean of 100 ms.

QRS complex
The normal full-term neonate has an axis between 55
and 200 but by 1 month, the normal upper limit has
fallen to 160 or less. Although one might identify an
axis of 120 as right axis deviation in an adult, it is a
normal finding in a newborn. The QRS axis in the
premature newborn ECG ranges between 65 and 174.
The duration of the QRS complex is measured from
the beginning to the end of the ventricular depolarization complex and it should be measured in a lead with
an initial Q wave[5]. QRS duration in the newborn and
infant is narrow (<80 ms). Normal QRS duration increases with age. Normal values for QRS complex
duration in lead V5 are displayed in Table 1.
QRS morphology in the newborn may have more
notches and direction changes than seen in older children or adults. The direction of the Q wave in the
precordial or horizontal plane indicates the direction of
septal depolarization. Normally, there is a Q wave in
leads V5V6 indicating depolarization from left to right.
Normal values of Q wave amplitudes vary with the lead
and with age. Q wave amplitudes may be as high

Task Force Report

as 055 mV in lead III or 033 mV in aVF at 1 month.

Q wave duration >30 ms is abnormal. The appearance
of secondary r waves (r or R ) in the right chest leads
is frequent in normal neonates.
Davignon et al.[3] provided normal values in infants.
The use of 2nd and 98th percentiles to define normality
implies that 4% of the population are abnormal for any
given single measurement, so normal ranges have to be
interpreted with caution (Table 1). Thomaidis et al.
published normal voltages from healthy term and
premature neonates[13].

1333

10 ms (1/4 of a mm) while we recognize that this may be

within measurement error.

ST segment and T wave

ST segment elevations >1 mm above the isoelectric line
are uncommon in the newborn. In neonates and infants
it is better to consider as the isoelectric line the TP
segment instead of the PQ segment. T waves are normally quite variable in the first week of life. After
1 week, the T wave is negative in lead V1 and positive
in V5V6.

QT interval
The QT interval is the interval between the beginning of
the QRS complex and the end of the T wave. The QT
measurement should be made in leads II, V5, and V6
with the longest value being used. The main diculty lies
in identifying correctly the point where the descending
limb of the T wave intersects the isoelectric line. Due to
the fast heart rate of infants the P wave may be
superimposed on the T wave, particularly when the QT
interval is prolonged. In this case, the end of the T wave
should be extrapolated by drawing a tangent to the
downslope of the T wave and considering its intersection
with the isoelectric line.
The QT interval duration changes with rate and it is
usually corrected (QTc) by using Bazetts formula. Correction of the QT interval requires a stable sinus rhythm
without sudden changes in the RR interval. QTc is equal
to QT interval in seconds divided by the square root of
the preceding RR interval in seconds. To avoid timeconsuming calculations, a simple chart (Fig. 1) where
the value of QTc is easily obtained by matching QT and
RR interval in millimetres (given the paper speed at
25 mm . s
1) has been produced. When heart rate is
particularly slow or fast the Bazetts formula may not be
accurate in the correction but it remains the standard for
clinical use.
The mean QTc on the 4th day of life is 40020 ms[1]
and, at variance with the adult, no gender dierences are
present[14]. Therefore, the upper normal limit of QTc (2
standard deviations above the mean, corresponding to
the 975 percentile) is 440 ms. By definition, 25% of
normal newborns are expected to have a QTc greater
than 440 ms. In healthy infants there is a physiological
prolongation of QTc by the second month (mean
410 ms) followed by a progressive decline[15], so that by
the sixth month QTc returns to the values recorded in
the first week.
Pitfalls with QT measurement. Despite its apparent
simplicity the measurement of the QT interval is fraught
with errors. The simple fact that a small square on the
ECG paper is equivalent to 40 ms explains why healthy
scepticism should accompany claims of clinical importance attached to very small degrees of QT prolongation. An attempt should be made to measure with

Abnormal electrocardiogram in the

newborn
Heart rate
Sinus arrhythmia
Since sinus arrhythmia is less pronounced at fast heart
rate, neonates show a more regular rhythm than young
children and adolescents, particularly in the first week of
life. Sinus arrhythmia should be dierentiated from
wandering pacemaker, which manifests itself with a
gradual change of P wave axis and morphology and that
is due to a shift of the pacemaker from the sinus node
to the atrium and the atrioventricular (AV) junction.
Although wandering pacemaker may accompany other
types of bradyarrhythmia, it has no pathologic meaning.
Work-up
No work-up should be necessary unless significant
bradycardia coexists.
Sinus tachycardia
Sinus tachycardia is a sinus rhythm with a heart rate
above the normal limit for age. In the newborn the
upper normal limit (98th percentile) is 166 beats . min
1
in the first week and 179 beats . min
1 in the first
month. After the sixth month the upper normal limit
declines to approximately 160 beats . min
1 and at
1 year is 151 beats . min
1. These values have been
measured from ECGs recorded when infants were awake
and quiet. It has to be noted that newborn infants may
transiently reach a heart rate up to 230 beats . min
1.
Causes. Sinus tachycardia may be a sign of any condition associated with an increase of cardiac output. The
most frequent causes of sinus tachycardia in the neonatal period are represented by fever, infection, anaemia, pain, and dehydration (hypovolaemia). Other
causes of sinus tachycardia include neonatal hyperthyroidism and myocarditis, particularly when it is not
proportionate to the level of fever. Myocarditis is
usually, but not necessarily, associated with other clinical signs, such as gallop rhythm, or ECG abnormalities,
including T wave changes and conduction disturbances.
Eur Heart J, Vol. 23, issue 17, September 2002

Figure 1 Chart for calculation of QTc. QTc, according to the Bazetts formula is obtained by matching QT and RR interval in millimetres, given the paper speed
at 25 mm . s
1. Corresponding values of RR interval and uncorrected QT interval are also indicated.

1334
Task Force Report

Eur Heart J, Vol. 23, issue 17, September 2002

Task Force Report

Finally, several drugs that are commonly used during

infancy, e.g. beta adrenergic agonists or theophyllin,
may induce sinus tachycardia. In the newborn, these
may have been transmitted across the placenta or
through breast milk.
Work-up
The evaluation of these patients should be performed
according to the underlying condition. If myocarditis is
suspected an echocardiogram should be performed. Appropriate acute treatment of causes of tachycardia may be
considered. Persistence of elevated rates should be further
evaluated.
Sinus bradycardia
Sinus bradycardia is defined as a sinus rhythm with
a heart rate below the normal limit. In the neonatal
period the lower normal limit (2nd percentile) is
91 beats . min
1 during the first week and
107 beats . min
1 in the first month of life. At the first
month the lower limit increases to 121 beats . min
1 and
declines to approximately 100 beats . min
1 in the following months. At 1 year the lower normal limit is
89 beats . min
1. These values apply to an ECG recorded in the awake state when heart rate is measured
over two respiratory cycles.
Causes. Central nervous system abnormalities, hypothermia, hypopituarism, increased intracranial pressure,
meningitis, drugs passed from the mother to infant,
obstructive jaundice, and typhoid fever represent causes
of sinus bradycardia. As a consequence when sinus
bradycardia is present on the surface ECG such conditions should be excluded. Hypothyroidism is another
cause of bradycardia and is often associated with the
so-called mosque sign, a dome-shaped symmetric T
wave in the absence of a ST segment. Transient sinus
bradycardia has been observed in newborns from antiRo/SSA positive mothers, especially women with lupus
erythematosus or other connective diseases.
A lower than normal heart rate has been described in
patients aected by LQTS, a phenomenon which is
evident in the neonatal period[16]. It may sometimes
represent the first sign of the disease during the foetal
period[17].
Work-up
24-h Holter monitoring may be helpful for further
evaluation when a heart rate below 8090 beats . min
1 is
present on surface ECG during infancy. Evaluation for
underlying conditions should be performed.
Other bradycardias
Sinus pauses in newborns may last from 800 to 1000 ms.
Pauses >2 s are abnormal. Sinus pauses may be followed
by escape beats which arise from the atria or from the
AV-junction. It has to be noted that even healthy
neonates may show periods of junctional rhythm, i.e. a
sequence of narrow QRS complexes in the absence of
preceding P waves.
Causes. Infants with autonomic nervous system dysfunction consisting of augmented vagal tone may have sinus

1335

bradycardia, or significant sinus pauses of several

seconds. These generally occur during feeding, sleep,
defecation, or other times of increased vagal tone.
Apparent life-threatening events (ALTE), described as
loss of consciousness accompanied by pallor and hypotonia, have been related to vagal overactivity which may
manifest as sinus pauses or abrupt bradycardia. ALTE
may be associated with apneic episodes, or gastroesophageal reflux, that may precede severe bradycardia.
Infants with LQTS not only tend to have sinus
bradycardia but may also have sinus pauses.
Work-up
24-h Holter monitoring may be useful for the assessment of significant bradycardia. Long pauses secondary to
excessive vagal tone may be eliminated by the use of
atropine, and rarely require pacemakers. Treatment of
other underlying diseases should be undertaken.

P wave
Abnormal P waves may be seen in infants with atrial
enlargement or non-sinus origin of the P wave. Ectopic
atrial rhythms originate most commonly from the low
right atrium (0 to
90), high left atrium (+90 to
+180) or the low left atrium (+180 to +270).
Right atrial enlargement and/or hypertrophy typically
produces increased P wave amplitude with a normal P
wave duration. The P wave axis usually remains normal
so the eect is usually best seen in lead II.
Left atrial enlargement and/or hypertrophy typically
produces an increased and prolonged negative terminal
deflection of the P wave in lead V1 (generally accepted as
>40 ms in duration and 01 mV in amplitude). Left atrial
enlargement also causes exaggerated notching of the P
wave in lead II although this is not a specific sign.
Work-up
An echocardiogram should be performed when clinically
indicated.

Atrioventricular conduction
During atrial tachycardia, it is possible to observe 1/1
conduction through the atrio-ventricular node at rates
over 300 beats . min
1.
Complete (third degree) atrioventricular block
Complete AV block implies complete absence of conduction from atrium to ventricle. ECG shows normal
atrial activation and slower dissociated regular QRS
complexes. Congenital complete block is observed in
complex congenital heart malformations[18].
Approximately one out of every 15 000 to 20 000 live
births results in a baby with isolated AV block. The
association between isolated neonatal AV block and
maternal connective tissue disease is well established and
ascribed to the presence of anti Ro/SSA and La-SSB
antibodies in the mothers. Nearly every mother with an
aected child has circulating antibodies. However, only
2 to 5% of women with known antibodies will have a
Eur Heart J, Vol. 23, issue 17, September 2002

1336

Task Force Report

first child with AV block[19]. Mortality rate in patients

with neonatal AV block is still high, especially during
the first 3 months of life[20]. Acquired complete AV
block is rare in neonates. It is mainly infective (viral
myocarditis, HIV infection) or may be related to
tumours.
First and second degree atrioventricular block
Neonates may present with first or second degree AV
block and rare reports exist demonstrating progression
to complete AV block after birth in children with and
without antibody mediated conduction disorders[21].
Long QT syndrome is occasionally complicated by
impaired atrioventricular conduction, mostly 2:1 AV
block[22,23]. Functional AV block can be observed in
neonates because they have a fast atrial rate and the P
wave falls within the very prolonged T wave. Cases of
infra-Hisian block location at the His-Purkinje level
have been demonstrated[24,25]. In spite of dierent treatment modes including the use of high doses of betablockers and pacing, there is still significant mortality.
Heart block associated with prolonged QT interval
has been described in neonates and infants receiving
cisapride. Second degree AV block due to QT interval
prolongation has been also reported with the use of
other agents such as diphemanil[26] or doxapram in
premature infants.
Work-up
In neonates and infants with AV conduction abnormalities clinical history of autoimmune disease and plasma
titres of maternal antibodies (anti Ro/SSA and antiLa/
SSB) should be performed. When neonates have abnormal
AV nodal conduction without maternal antibodies, an
ECG should also be performed on the parents and siblings
(see intraventricular abnormalities). Neonates with first
degree AV block should be followed with additional ECGs
in the following months. Neonates and infants with second
or third degree AV block need a complete paediatric
cardiologic work-up, including an echocardiogram. The
only eective treatment of congenital complete AV block
in neonates with symptoms or a low ventricular escape
rhythm is permanent artificial pacing.

Intraventricular conduction
Bundle branch block
Congenital isolated complete right (RBBB) and left
bundle branch block are very rare in neonates. Southall
et al. found only one case of complete RBBB in a
population of 3383 apparently healthy newborn infants[27]. The classical ECG in Ebsteins anomaly of the
tricuspid valve displays a prolonged PR interval and a
wide RBBB. Left anterior fascicular block is found in
association with congenital heart malformations such as
atrio-ventricular canal defects and tricuspid atresia. In
severe cardiomyopathy, interruption of the left bundle,
which results from the involvement of the left ventricle
and/or its conduction system, has been reported and
carries a poor prognosis[28].
Eur Heart J, Vol. 23, issue 17, September 2002

Hereditary bundle branch block is an autosomal

dominant genetic disease that was mapped in some
families to the long arm of chromosome 19[29,30]. Affected individuals have various combinations of conduction defects such as RBBB, left or right QRS axis
deviation or AV block; the r pattern may as well be the
prelude to a conduction block. Abnormalities have been
described in patients as young as 15 days.
Non-specific intraventricular conduction abnormalities
Non-specific intraventricular conduction abnormalities
are very rare in neonates and infants with normal heart
structures[31]. They may be a manifestation of inflammation in myocarditis or endocarditis.
Work-up
Neonates and infants with intraventricular conduction
abnormalities need a complete paediatric cardiologic
work-up. Evaluation of possible underlying causes should
be performed. An ECG should also be performed on the
parents and siblings.
WolParkinsonWhite syndrome
The anatomical substrate of preexcitation in Wol
ParkinsonWhite (WPW) syndrome is a direct muscular
connection between the atria and ventricles. Since accessory pathways rarely show decremental conduction, the
electrical impulse is conducted prematurely to the ventricles resulting in a short PR interval. Conduction
through the atrioventricular node and the accessory
pathway results in collision of two electrical wavefronts
at the ventricular level causing a delta wave and a fusion
QRS complex with prolonged duration.
The diagnosis of preexcitation is solely based on the
findings of the surface ECG (Fig. 2). Intermittent preexcitation is not uncommon in newborns and infants.
Depending on the location of the accessory pathway as
well as the conduction properties of the atrioventricular
node, even continuous preexcitation may be subtle and
only detected in the mid-precordial leads. A study in
newborns indicated a high prevalence of WPW syndrome when two of the four following characteristics
were noted: PR interval c100 ms, QRS complex duration d80 ms, lack of a Q wave in V6 and left axis
deviation[32]. Short PR intervals are also observed in
mannosidosis, Fabrys disease, and Pompes disease[5]. A
common cause of a short PR interval in a normal heart
is a low right atrial pacemaker. In this instance, the P
wave is negative in lead aVF and positive or isoelectric
in lead I. The intraatrial conduction time from the high
to low right atrium is eliminated and therefore the PR
interval may be up to 40 ms less than normal.
The prevalence of WPW syndrome in the paediatric
population has been estimated at 015 to 03%[33] with an
incidence of newly diagnosed cases of approximately
four per 100 000 persons per year for all age groups[34].
Numbers, however, vary greatly depending upon symptoms, age, gender and the intracardiac anatomy of the
population studied[35,36]. In children with structural
heart disease, the prevalence has been estimated at 033
to 05%[37]. Ebsteins anomaly of the tricuspid valve,

Figure 2

ECG in a neonate showing subtle signs of WPW. Note delta waves in V5 and V6 with absence of Q waves.

Task Force Report

1337

Eur Heart J, Vol. 23, issue 17, September 2002

1338

Task Force Report

l-transposition of the great arteries, hypertrophic cardiomyopathy and cardiac tumours are associated with an
increased prevalence of preexcitation[36,37].
Clinical counterparts. In WPW syndrome the typical
form of paroxysmal supraventricular tachycardia (orthodromic) results from reentry antegradely through the
atrioventricular node and retrogradely through the accessory pathway. As digoxin shortens the antegrade
eective refractory period of the accessory pathway and
promotes rapid atrioventricular conduction during atrial
flutter or atrial fibrillation over the pathway, the use of
digoxin is contraindicated at any age [38,39]. Verapamil
should also be avoided as it may increase the ventricular
response rate during atrial fibrillation in those patients,
and may cause cardiovascular collapse in infants and
young children.
The incidence of sudden death in preexcitation syndrome during childhood has been estimated to be as
high as 05%[34] and cardiac arrest may be the initial
presentation in children with preexcitation[35]. However,
data on newborns and infants are lacking. One study on
a series of 90 newborns and infants with WPW syndrome and supraventricular tachycardia reported sudden death in two patients with a normal heart during
follow-up. Both infants, however, had been treated with
digoxin[36].
Finally, there are no sucient data on newborns and
infants with an incidental finding of preexcitation on
ECG concerning the occurrence of paroxysmal supraventricular tachycardia later on during their life.
Work-up
Congenital heart disease is more common in infants and
young children with preexcitation, with a prevalence as
high as 45% for infants with an ECG pattern consistent
with a right-sided accessory pathway[36]. Thus, in every
young patient with a preexcitation pattern on surface
ECG, a complete 2-dimensional echocardiographic
work-up is recommended to rule out any intracardiac
abnormality.
Assessment of the conduction properties of the accessory pathway, i.e. the antegrade eective refractory period
and the shortest RR-interval with preexcitation, by transesophageal programmed stimulation may be useful in
selected patients for risk stratification and mode of
therapy.

QRS axis and amplitude

Abnormal axis implies a mean frontal plane QRS vector
outside the normal range and must take into account the
relative right axis deviation seen in normal neonates. Left
axis deviation is seen in a variety of abnormalities including atrioventricular septal defect, ventricular septal defect, tricuspid atresia, and WPW syndrome, but may be
occasionally observed in otherwise normal infants.
Right ventricular hypertrophy
Right ventricular hypertrophy may be suspected from a
QR complex in V1, an upright T wave in V1 (normal in
Eur Heart J, Vol. 23, issue 17, September 2002

the first week of life), increased R wave amplitude in V1,

and increased S wave amplitude in V6 (according to the
Davignon criteria). Sensitivity and specificity has not
been tested in the neonate. QR patterns are commonly
seen with pressure overload congenital lesions, rSR
patterns are seen in volume overload lesions.
Left ventricular hypertrophy
The performance of the ECG in recognition of left
ventricular hypertrophy is poorer than generally recognized and, again, has not been specifically tested in
neonates. Left ventricular hypertrophy is expected to
produce increased left sided voltages. Garson[5] described the most helpful ECG signs in children as being
T wave abnormalities in leads V5 and V6, increased R
wave amplitude in V6, increased S wave amplitude in V1
(according to the Davignon criteria), and a combination
of these last two variables. Left to right shunt lesions
may result in left ventricular hypertrophy, but this may
be in association with right ventricular hypertrophy and
manifested as biventricular hypertrophy. Left ventricular hypertrophy in the newborn may be attenuated by
the normal right-sided predominance of the newborn.
The normal premature heart may not have developed
the right-sided predominance, especially if <28 weeks
gestation, and left ventricular predominance may be
present.
Low QRS voltage
In the limb leads the total amplitude of R+S in each
lead c05 mV may be indicative of myocarditis or
cardiomyopathy.
Work-up
Evaluation of the underlying causes should be performed. An echocardiogram should be performed when
clinically indicated.

Ventricular repolarization
There is a simple reason that makes clinically important
the analysis of ventricular repolarization abnormalities:
their presence could be the harbinger of a significant risk
for life-threatening arrhythmia. It is established that
newborns found to have a prolonged QTc (>440 ms) on
the fourth day of life have an increased risk for sudden
death[1]. Some of these sudden deaths have previously
been labelled as Sudden Infant Death Syndrome.
On the other hand, the presence of confounding
factors above all the ambiguities in their
quantification calls for caution before making hasty
diagnoses associated with need for therapy and with
considerable parental anxiety.
Ventricular repolarization can be evaluated on the
surface ECG by measuring the QT interval duration and
by analysing the morphology of the ST segment and of
the T wave. Measurements of the QT interval should be
performed by hand.
It is important to remember that QT duration may
change over time. Accordingly, it is recommended

Task Force Report

1339

Figure 3 ECG tracings of three newborns with LQTS diagnosed in the first months of life. Mutations on the
potassium channel gene KvLQT1 (panel A) and on the sodium channel gene SCN5A (panel B) were identified. Panel
B modified from ref. [44].

repeating the ECG in those infants found to have a

prolonged QTc on the first ECG. While exceptions do
exist, the more prolonged the QTc interval, the greater
the likelihood of its clinical significance. A QTc close to
500 ms implies a clear abnormality even taking into
account potential measurement errors.
QT interval prolongation: dierential diagnosis
Electrolyte disturbances are fairly common and may
cause QT prolongation. Among them, hypocalcaemia
(less than 75 mg . dl
1) usually produces a distinctive
lengthening of the ST segment. Hypokalaemia and
hypomagnesaemia, often encountered in infants who
have had vomiting or diarrhoea, usually decrease T
wave amplitude and increase U wave amplitude. Central
nervous system abnormalities can produce QT prolongation and T wave inversion.
Several drugs commonly used in the neonatal period
and during infancy may induce QT interval prolongation; among them are macrolide antibiotics such as
spyramycin[40], erythromycin, clarithromycin and also
trimethoprim. Prokinetics such as cisapride have been
positively linked to QT interval prolongation. All these
drugs share one action: they block IKr, one of the
ionic currents involved in the control of ventricular
repolarization.
Neonates born from mothers with autoimmune diseases and positive for the anti-Ro/SSA antibodies may
also show QT interval prolongation, sometimes with
QTc values exceeding 500 ms[41], which tends to be

transient and to disappear by the sixth month of life,

concomitantly with the disappearance of the anti Ro/
SSA antibodies.
Finally, some of the neonates with QT interval prolongation may be aected by the congenital LQTS. This
possibility has to be carefully evaluated because of its
implications for management.
Long QT syndrome
Long QT syndrome, whose prevalence appears to be
close to 1/30001/5000, is characterized by the occurrence of syncopal episodes due to torsades de pointes
ventricular tachycardia (VT) and by a high risk for
sudden cardiac death among untreated patients[42]. Importantly, in 12% of patients with LQTS, sudden death
was the first manifestation of the disease and in 4% this
happened in the first year of life[42]. This point alone
mandates the treatment of all those diagnosed as affected, even if there are no symptoms. Long QT syndrome is a genetic disease due to mutations of several
genes all encoding ionic (potassium or sodium) currents
involved in the control of ventricular repolarization. In
most cases, several members of the same family are
gene-carriers. Low penetrance exists in LQTS, which
means that gene-carriers may not show the clinical
phenotype and may have a normal QT interval[43].
Therefore a normal QT in the parents does not rule
out familial LQTS. In addition, approximately 30% of
cases are due to de novo mutations which imply
unaected parents and no family history. De novo
Eur Heart J, Vol. 23, issue 17, September 2002

1340

Task Force Report

Figure 4 QT prolongation management flow chart. Electrolytes, echocardiogram, intracranial ultrasound are recommended in the appropriate clinical situation. In cases of positive genetics and QTc >440 ms, therapy is indicated. #In
cases of a positive family history (Hx) for LQTS; *See text.

LQTS mutations have been demonstrated in infant

victims of cardiac arrest and sudden death diagnosed as
Sudden Infant Death Syndrome[2,44,45]. Even though
relatively few LQTS patients have cardiac events during
the first year of life, the vast majority become symptomatic later on, either during childhood or adolescence
according to genetic subgroups[46]. Therefore treatment
must continue. Beta-blockers are the first choice therapy
in LQTS and are eective in preventing recurrences in
80% of already symptomatic patients; dierent degrees
of protection exist according to genetic subgroups[46]. If
beta-blockers are unable to prevent new cardiac events,
additional drug therapy, left cardiac sympathetic denervation, pacemakers or the implantable cardioverter
defibrillator should be considered based on evidence,
with due consideration for body size.
ECG tracings of newborns with LQTS are shown
in Fig. 3.
Work-up
It is well understood that the likelihood of having LQTS
increases with increasing QTc; however, since a small
percentage of LQTS patients has a QTc <440 ms[47], the
correlation between QT prolongation and the presence of
the syndrome is not absolute. Therefore, the following discussion is presented as guidelines based upon
experience and current knowledge, and is likely to be
updated frequently. Given the life-threatening potential of
the disease, once the diagnosis of LQTS becomes probable, it is recommended that these infants are referred to a
specialist as soon as possible.
 First ECG: QTc above 440 ms, the upper limit of
normal.
Exclude other causes of acquired QT interval prolongation
and obtain a detailed family history for the possibility of
familial LQTS. Episodes of early sudden death, fainting
spells, and seizures-epilepsy should alert to this possibility.
The ECG should be repeated after a few days to confirm
the abnormal finding. Subsequent management depends on
(1) presence or absence of family history suggestive for
LQTS, and (2) the degree of QT interval prolongation.
The presence of complex ventricular arrhythmias would
Eur Heart J, Vol. 23, issue 17, September 2002

have additional importance. The following stepwise approach involves infants with and without a family history
for LQTS (Fig. 4). If family history is positive, then as
LQTS is an autosomal dominant disease the infant has
a 50% probability of being aected and complete diagnostic procedures should be performed, as always with LQTS
families.
 The second ECG is normal.
If the first QTc was <470 ms, dismiss the case. If the
first QTc was d470 ms, then plan a third ECG after
12 months to remain on the safe side.
 The second ECG shows a QTc between 440 and 470 ms.
In these cases with persistent borderline QT prolongation,
electrolytes, including calcium and magnesium, should be
checked. Clinical history of autoimmune disease and
plasma titres of maternal antibodies (anti Ro/SSA and
antiLa) should be performed. T wave morphology may be
helpful; for example, the presence of notches on the T
wave in the precordial leads further suggests the presence
of LQTS[48]. Additionally, mild bradycardia can also be
found in LQTS. ECGs should be obtained from the
parents and siblings of the neonate.
In the absence of family history of LQTS, symptoms
or arrhythmias, a 24-h Holter monitoring should be
obtained to look for T wave alternans, complex ventricular arrhythmias or marked QTc prolongation, and the
ECG should be periodically checked during the first
year. No treatment is currently recommended. With a
positive family history, the probability of LQTS becomes
high. Additional diagnostic procedures (24-h Holter
monitoring, echocardiogram and genetic screening)
should be performed and initiation of therapy could be
considered.
 The second ECG shows a QTc d470 and <500 ms.
All diagnostic procedures listed above should be performed
and a third ECG should be planned within a month. In
case of a positive family history, therapy should be
initiated. Even without a family history, therapy should be
considered.
Even in infants with a very prolonged QTc in the first
month of life, the ECG may normalize. If subsequent

Task Force Report

1341

Figure 5 ECG tracings of a newborn with blocked atrial bigeminy which simulates sinus bradycardia. Blocked
P waves are present when examining the T waves.
ECGs and diagnostic procedures do not confirm the
presence of LQTS, it is logical to progressively withdraw
therapy and to return to periodic observations.
 The second ECG shows a QTc d500 ms.
Infants with a QTc d500 ms are very likely to be aected
by LQTS and to become symptomatic. All diagnostic
procedures listed above should be performed and these
infants should be treated.
Highest risk. The presence of QTc close to 600 ms, or of
T wave alternans, or of 2:1 AV block secondary to major
QT prolongation, or of hearing loss, identify infants at
extremely high risk[47,49].
ST segment elevation
Causes. There are multiple causes of ST segment elevation in infancy. The most frequent is pericarditis. Less
frequent causes of ST segment elevation with or without
T wave abnormalities are hyperkalaemia, intracranial
haemorrhage, pneumothorax and pneumopericardium,
subepicardial injury due to anomalous left coronary
artery or to Kawasaki disease with cardiac involvement.
Whenever an anomalous left coronary artery or, more
frequently, Kawasaki disease produce an acute myocardial infarction, QT prolongation may also be present.
ST segment elevation with a RBBB pattern in the
right precordial leads (V1V2) is the typical finding of the
Brugada syndrome, a genetic disorder associated with a
high incidence of sudden cardiac death secondary to
ventricular fibrillation, in the absence of cardiac structural abnormalities[50]. The ST segment elevation is
typically downsloping or coved and it is followed by a
negative T wave, at variance with the early repolariz-

ation syndrome where ST segment elevation has an

upward concavity, it is confined to mid-precordial leads
and is associated with a positive T wave. The diagnosis
of the Brugada syndrome is made dicult by the intermittent nature of the ECG abnormalities, as 40% of
cases may normalize transiently. Rare cases of Brugada
syndrome have been reported during infancy[51,52].
Work-up
Whenever the underlying cause has been identified, it
should be treated. If the Brugada syndrome is suspected,
careful family history should be collected, 24-h Holter
monitoring obtained, and the patient should be referred to
a specialist.

Atrial and ventricular arrhythmias

Atrial/junctional
Premature atrial beats. A premature atrial beat is a
premature P wave. Premature atrial beats usually have
a dierent morphology and mean vector from sinus
P waves. In regular sinus rhythm at a normal rate, a P
wave that occurs before the next expected P wave is a
premature atrial beat. A premature atrial beat may be
conducted to the ventricles normally, or with ventricular
aberration or not conducted or blocked. Occasionally,
blocked premature atrial beats occur in a bigeminal
sequence, so-called blocked atrial bigeminy. This
rhythm simulates sinus bradycardia; it is important to
examine the T waves carefully for blocked P waves (Fig.
5). In infants, since the refractory periods of the bundle
branches are similar, premature atrial beats may be
conducted with either RBBB or left bundle branch
Eur Heart J, Vol. 23, issue 17, September 2002

1342

Task Force Report

Table 2

Distinguishing tachyarrhythmias in infants

Sinus tachycardia

SVT

Atrial flutter

VT

History

Sepsis, fever, hypovolaemia,

etc.

Usually otherwise normal

Most have a normal heart

Many with abnormal

heart

Rate

Almost always <230 b/min

Most often 260300 b/

min

Atrial 300500 b/min. Vent.

1:1 to 4:1 conduction

200500 b/min

R-R interval
variation

Over several seconds may get

faster and slower

After first 1020 beats,

extremely regular

May have variable block

(1:1, 2:1, 3:1) giving dierent
ventricular rates

Slight variation over

several beats

P wave axis

Same as sinus almost always

visible P waves

60% visible P waves,

P waves do not look
like sinus P waves

Flutter waves (best seen in

LII, LIII, aVF, V1)

May have sinus P waves

continuing unrelated to
VT (AV dissociation),
retrograde P waves, or
no visible P waves

QRS

Almost always same as

slower sinus rhythm

After first 1020 beats,

almost always same as
sinus

Usually same as sinus, may

have occasional beats
dierent from sinus

Dierent from sinus (not

necessarily wide)

SVT=Supraventricular tachycardia.
VT=ventricular tachycardia.

aberration. It is relatively common in the same strip

to see premature atrial beats conducted normally,
aberrantly and blocked. It is relatively uncommon for
an infant to have both premature atrial beats and
premature ventricular beats, although it does occur.
Therefore, in an infant with premature P waves and wide
QRS complexes on the same strip, a careful search
should be made for a premature P wave preceding a
wide QRS, before making the diagnosis of both premature atrial beats and premature ventricular beats on the
same strip.
Work-up
In patients with frequent premature atrial beats, a
follow-up ECG at 1 month may be performed. Relatively
long periods of blocked atrial bigeminy may simulate sinus
bradycardia. The distinction is important since blocked
atrial bigeminy is most often benign while severe sinus
bradycardia may accompany systemic illness.
Supraventricular tachycardia. Supraventricular tachycardia (SVT) is a rapid regular tachyarrhythmia, which
results from an abnormal mechanism originating proximal to the bifurcation of the bundle of His and does not
have the morphology of atrial flutter. This definition of
SVT specifically excludes sinus tachycardia (Table 2)
and premature wide QRS. The usual infant with SVT
has an extremely regular R-R interval after the first
1020 beats, most often at rates greater than 230
beats . min
1 and usually 260300 per min. In 60% of
cases, the P waves are visible, but the P waves almost
always have a dierent morphology from sinus. In over
90% of infants and children with SVT, the QRS complex
is narrow and in only 3% is the QRS complex dierent
from the underlying sinus QRS. Therefore, the persistent
aberration of SVT in infants is exceedingly rare, implying that in the majority of infants with a QRS complex
dierent from sinus, the diagnosis is VT. In very rare
Eur Heart J, Vol. 23, issue 17, September 2002

cases of SVT in infants, there may be atrial tachycardia

with AV block or junctional tachycardia with AV
dissociation.
Work-up
It is important to document SVT with a 12-lead ECG
before attempting conversion of the rhythm unless the
infant is critically ill. After sinus rhythm is achieved, the
WPW pattern should be sought on a 12-lead ECG.
Treatment to prevent further episodes of SVT in infancy is
generally recommended. An echocardiogram is indicated
to determine ventricular function or the presence of
congenital heart disease.
Atrial flutter. Atrial flutter is characterized by a rapid,
regular form of atrial depolarization: the flutter wave.
The picket fence morphology is similar to adults. However, the flutter wave durations are generally 009 to
018 s with atrial rates in infants between 300500
beats . min
1. In general, there is variable AV conduction from 1:1 to 4:1 yielding an irregular ventricular rate.
The QRS complex is usually the same as in sinus rhythm
although there may be occasional aberrancy (Table 2).
Due to the occasional association with WPW, this
pattern should be specifically sought.
Other types of supraventricular arrhythmias such as
atrial fibrillation or multifocal tachycardia are extremely
rare in the neonate.
Work-up
Conversion to sinus rhythm should be attempted. An
echocardiogram is worthwhile to determine ventricular function and the possible presence of congenital heart disease.
Ventricular arrhythmias
Premature ventricular beats. A premature ventricular
beat is manifest on the surface ECG as a premature
abnormal QRS (not similar to the sinus QRS complex)
that is not preceded by a premature P wave. It is

Task Force Report

important to recognize that in infants, the QRS duration

may be normal or slightly prolonged (i.e. less than
008 s) but if the complex has a dierent morphology
from the sinus, and is not preceded by a premature P
wave, the diagnosis is a premature ventricular beat.
The relationship between morphology and the site of
origin is not exact enough to be able to predict which
ventricle is causing the arrhythmia. It is not possible to
distinguish premature ventricular beats from premature
atrial beats with aberrancy on the basis of QRS
morphology.
Work-up
The QT interval should be measured carefully during
periods of sinus rhythm (see section on repolarization abnormalities). In complex ventricular arrhythmias,
24-h Holter monitoring may be worthwhile. An echocardiogram may be performed to determine ventricular
function or structural abnormalities. Occasionally maternal drugs that cause ventricular arrhythmias may be
transferred in utero or post-natally in breast milk.
Ventricular tachycardia. Ventricular tachycardia is a
series of three or more repetitive complexes that originate from the ventricles. The complexes are therefore
dierent from the patients normal QRS; usually, the
QRS duration is prolonged for the age of the patient
(009 s or more in infants). It is therefore usually a wide
QRS tachycardia. However, infants may have a QRS
duration in VT less than 009 s but clearly dierent from
the sinus complex (Table 2). The specific morphology of
the QRS is generally not helpful to distinguish VT from
supraventricular tachycardia with aberration. However,
SVT in infants with a dierent QRS beyond the first
1020 beats is rare, therefore, in this situation, a diagnosis of VT should be strongly considered. In infants,
the rate of VT may be from 200500 beats . min
1.
There may be a slight variation in the R-R interval over
several beats. There may be sinus P waves continuing
unrelated to VT (AV dissociation), retrograde P waves
or no visible P waves. There may also be fusion and
capture beats. The diagnosis of VT should be strongly
considered if the patient has premature ventricular beats
during times of sinus rhythm with a similar morphology
to the tachyarrhythmia.
Work-up
An underlying cardiac or central nervous system
abnormality may be found in infants with VT. The QT
interval should be carefully measured (see section on
repolarization abnormalities), 24-h Holter monitoring and
echocardiogram should be obtained. Treatment is generally indicated.
Accelerated ventricular rhythm. Accelerated ventricular
rhythm is also known as slow VT generally the rate
is <200 beats . min
1. It occurs at approximately the
same rate as the infants sinus rate, and the rhythms tend
to alternate.
Work-up
While these infants most often have a normal heart, a
work-up similar to VT is indicated.

1343

Conclusion
The main objective of the present document was to
provide cardiologists, who work with adults, with a
practical approach to neonatal electrocardiography, and
paediatricians and neonatologists with a tool that should
facilitate medical interaction on cardiologic issues. There
are important dierences between neonatal and adult
ECGs. When a cardiologist examines the ECG of an
apparently normal and healthy infant the focus has to be
on distinguishing between patterns that should cause no
alarm and those that require action or additional investigations. To provide clues for this distinction is what the
members of the Task Force have attempted to do and,
whenever possible or appropriate, steps in management
have also been suggested. This document is not intended
to be all-inclusive or to substitute for textbooks on
paediatric and neonatal ECG.
The manuscript has been reviewed by the Members of
the CPGPC: Werner Klein (Chair), Maria Angeles Alonso,
Gianfranco Mazzotta, Carina Blomstrom Lundqvist. The authors
are grateful to Pinuccia De Tomasi, BS, for expert editorial support
in the preparation of the manuscript.

References
[1] Schwartz PJ, Stramba-Badiale M, Segantini A et al. Prolongation of the QT interval and the sudden infant death
syndrome. N Engl J Med 1998; 338: 170914.
[2] Schwartz PJ, Priori SG, Bloise R et al. Molecular diagnosis in
a child with sudden infant death syndrome. Lancet 2001; 358:
13423.
[3] Davignon A, Rautaharju P, Boisselle E, Soumis F, Megelas
M, Choquette A. Normal ECG standards for infants and
children. Pediatr Cardiol 1979; 1: 12352.
[4] Rijnbeek PR, Witsenburg M, Schrama E, Hess J, Kors JA.
New normal limits for the paediatric electrocardiogram. Eur
Heart J 2001; 22: 70211.
[5] Garson A Jr. Electrocardiography. In: Garson A Jr, Bricker
JT, Fisher DJ, Neish SR, eds. The Science and Practice of
Paediatric Cardiology, 2nd edn. Baltimore, MD: Williams &
Wilkins, 1998: 71388.
[6] Tippel M. Interpretation of electrocardiograms in infants
and children. Images Paediatr Cardiol 1999; 1: 313 available
at:
http://www.health.gov.mt/impaedcard/issue/issue1/ecg1/
ect1.htm#top
[7] Garson A Jr. Clinically significant dierences between the
old analog and the new digital electrocardiograms. Am
Heart J 1987; 114 (1 Pt 1): 1947.
[8] Rijnbeek PR, Kors JA, Witsenburg M. Minimum bandwidth
requirements for recording of paediatric electrocardiograms.
Circulation 2001; 104: 308790.
[9] Yamamoto H, Miyahara H, Domae A. Is a higher sampling
rate desirable in the computer processing of the paediatric
electrocardiogram? J Electrocardiol 1987; 20: 3218.
[10] Macfarlane PW, Coleman EN, Pomphrey EO, McLaughlin S,
Houston A, Aitchison T. Normal limits of the high-fidelity
paediatric ECG. Preliminary observations. J Electrocardiol
1989; 22 (Suppl): 1628.
[11] Bailey JJ, Berson AS, Garson A Jr et al. Recommendations
for standardization and specifications in automated electrocardiography: bandwidth and digital signal processing. A
report for health professionals by an ad hoc writing group of
the Committee on Electrocardiography and Cardiac Electrophysiology of the Council on Clinical Cardiology, American
Heart Association. Circulation 1990; 81: 7309.
Eur Heart J, Vol. 23, issue 17, September 2002

1344

Task Force Report

[12] Stramba-Badiale M, Lazzarotti M, Schwartz PJ. Development of cardiac innervation, ventricular fibrillation, and
sudden infant death syndrome. Am J Physiol 1992; 263:
H151422.
[13] Thomaidis C, Varlamis G, Karamperis S. Comparative study
of the electrocardiograms of healthy fullterm and premature
newborns. Acta Paediatr Scand 1988; 77: 6537.
[14] Stramba-Badiale M, Spagnolo D, Bosi G, Schwartz PJ. Are
gender dierences in QTc present at birth? MISNES Investigators. Multicenter Italian Study on Neonatal Electrocardiography and Sudden Infant Death Syndrome. Am J Cardiol
1995; 75: 12778.
[15] Schwartz PJ, Montemerlo M, Facchini M et al. The QT
interval throughout the first 6 months of life: a prospective
study. Circulation 1982; 66: 496501.
[16] Vincent GM. Heart rate of RomanoWard syndrome
patients. Am Heart J 1986; 112: 614.
[17] Hofbeck M, Ulmer H, Beinder E, Sieber E, Singer H. Prenatal
findings in patients with prolonged QT interval in the neonatal
period. Heart 1997; 77: 198204.
[18] Ho SY, Fagg N, Anderson RH, Cook A, Allan L. Disposition
of the atrioventricular conduction tissues in the heart with
isomerism of the atrial appendages: its relation to congenital
complete heart block. J Am Coll Cardiol 1992; 20: 90410.
[19] Brucato A, Frassi M, Franceschini F et al. Risk of congenital
complete heart block in newborns of mothers with anti-Ro/
SSA antibodies detected by counterimmunoelectrophoresis: a
prospective study of 100 women. Arthritis Rheum 2001; 44:
18325.
[20] Buyon JP, Hiebert R, Copel J et al. Autoimmune-associated
congenital heart block: long-term outcome of children and
immunogenetic study. J Am Coll Cardiol 1998; 31: 165866.
[21] Geggel RL, Tucker L, Szer I. Postnatal progression from
second to third degree heart block in neonatal lupus syndrome. J Pediatr 1988; 113: 104952.
[22] Trippel DL, Parsons MK, Gillette PC. Infants with long-QT
syndrome and 2:1 atrio-ventricular block. Am Heart J 1995;
130: 11304.
[23] Gorgels APM, Al Fadley F, Zaman L, Kantoch MJ, Al
Halees Z. The long QT syndrome with impaired atrioventricular conduction: a malignant variant in infants. J Cardiovasc
Electrophysiol 1998; 9: 122532.
[24] Van Hare GF, Franz MR, Roge C, Scheinman MM. Persistent functional atrio-ventricular block in two patients with
prolonged QT intervals: elucidation of the mechanism of
block. PACE 1990; 13: 60818.
[25] Lupoglazo JM, Cheav T, Baroudi G et al. Homozygous
SCN5A mutation in long QT syndrome with functional
two-to-one atrioventricular block. Circ Res 2001; 89: e16e21.
[26] Villain E, Kachaner J, Le Bidois J et al. Bloc auricoloventriculaire partiel et allongement de QT chez quatre prematures recevant du diphemanil. Arch Fr Pediatr 1990; 47: 335.
[27] Southall DP, Johnson AM, Shinebourne EA, Johnston PG,
Vulliamy DG. Frequency and outcome of disorders of cardiac
rhythm and conduction in a population of newborn infants.
Paediatrics 1981; 68: 5866.
[28] Gnota JF, Samson RA. Left bundle branch block in infants with dilated cardiomyopathy conveys a poor prognosis.
Cardiol Young 1999; 9: 557.
[29] Stephan E, de Meeus A, Bouvagnet P. Hereditary bundle
branch defect: right bundle branch blocks of dierent causes
have dierent morphologic characteristics. Am Heart J 1997;
133: 24956.
[30] Brink PA, Ferreira A, Moolman JC, Weymar HW, van der
Merwe PL, Corfield VA. Gene for progressive familial heart
block type I maps to chromosome 19q13. Circulation 1995;
91: 163340.
[31] Ewing LL. Intraventricular conduction disturbances in children. Prog Pediatr Cardiol 1994; 4: 1119.
[32] Perry JC, Giure RM, Garson A. Clues to the electrocardiographic diagnosis of subtle WolParkinsonWhite syndrome. J Pediatr 1990; 117: 8715.
Eur Heart J, Vol. 23, issue 17, September 2002

[33] Sorbo MD, Buja GF, Miorelli M et al. The prevalence of the
WolParkinsonWhite syndrome in a population of 116 542
young males. G Ital Cardiol 1995; 25: 6817.
[34] Munger TM, Packer DL, Hammill SC et al. A population
study of the natural history of WolParkinsonWhite syndrome in Olomsted County, Minnesota, 19531989. Circulation 1993; 87: 86673.
[35] Paul T, Guccione P, Garson A Jr. Relation of syncope in
young patients with WolParkinsonWhite syndrome to
rapid ventricular response during atrial fibrillation. Am J
Cardiol 1990; 65: 31821.
[36] Deal BJ, Keane JF, Gillette PC, Garson A. WolParkinson
White syndrome and supraventricular tachycardia during
infancy: management and follow-up. J Am Coll Cardiol 1985;
5: 1305.
[37] Perry JC, Garson A. Supraventricular tachycardia due to
WolParkinsonWhite syndrome in children: early disappearance and late recurrence. J Am Coll Cardiol 1990; 16:
121520.
[38] Saul JP, Walsh EP, Triedman JK. Mechanisms and therapy of
complex arrhythmias in paediatric patients. J Cardiovasc
Electrophysiol 1995; 6: 112948.
[39] Villain E, Bonnet D, Acar P, Aggoun Y, Sidi D, Kachaner J.
Recommendations for the treatment of recurrent supraventricular tachycardia in infants. Arch Pediatr 1998; 5:
1338.
[40] Stramba-Badiale M, Nador F, Porta N et al. QT interval
prolongation and risk of life-threatening arrhythmias during
toxoplasmosis prophylaxis with spiramycin in neonates. Am
Heart J 1997; 133: 10811.
[41] Cimaz R, Stramba-Badiale M, Brucato A, Catelli L, Panzeri
P, Meroni PL. QT interval prolongation in asymptomatic
anti-SSA/Ro-positive infants without congenital heart block.
Arthritis Rheum 2000; 43: 104953.
[42] Schwartz PJ, Priori SG, Napolitano C. The long QT syndrome. In: Zipes DP, Jalife J, eds. Cardiac electrophysiology:
from cell to bedside, 3rd edn. Philadelphia: WB Saunders,
2000: 597615.
[43] Priori SG, Napolitano C, Schwartz PJ. Low penetrance in the
long QT syndrome. Clinical impact. Circulation 1999; 99:
52933.
[44] Schwartz PJ, Priori SG, Dumaine R et al. A molecular link
between the sudden infant death syndrome and the long-QT
syndrome. N Engl J Med 2000; 343: 2627.
[45] Ackerman MJ, Siu BL, Sturner WQ et al. Postmortem
molecular analysis of SCN5A defects in sudden infant death
syndrome. JAMA 2001; 286: 22649.
[46] Schwartz PJ, Priori SG, Spazzolini C et al. Genotypephenotype correlation in the long-QT syndrome: gene-specific
triggers for life threatening arrhythmias. Circulation 2001;
103: 8995.
[47] Garson A Jr, Dick M 2nd, Fournier A et al. The long QT
syndrome in children. An international study of 287 patients.
Circulation 1993; 87: 186672.
[48] Malfatto G, Beria G, Sala S, Bonazzi O, Schwartz PJ.
Quantitative analysis of T wave abnormalities and their
prognostic implications in the idiopathic long QT syndrome.
J Am Coll Cardiol 1994; 23: 296301.
[49] Villain E, Levy M, Kachaner J, Garson A Jr. Prolonged QT
interval in neonates: benign, transient, or prolonged risk of
sudden death. Am Heart J 1992; 124: 1947.
[50] Brugada J, Brugada R, Brugada P. Right bundle branch block
and ST-segment elevation in leads V1 through V3: a marker
for sudden death in patients without demonstrable structural
heart disease. Circulation 1998; 97: 45760.
[51] Priori SG, Napolitano C, Giordano U, Collisani G, Memmi
M. Brugada syndrome and sudden cardiac death in children.
Lancet 2000; 355: 8089.
[52] Suzuki H, Torigoe K, Numata O, Yazaki S. Infant case with
a malignant form of Brugada Syndrome. J Cardiovasc
Electrophysiol 2000; 11: 127780.